Respiratory uncoupling by increased H⁺ or K⁺ flux is beneficial for heart mitochondrial turnover of reactive oxygen species but not for permeability transition

Hearts exposed to brief periods of ischemia exhibit considerable protection from subsequent ischemia reperfusion-injury, so-called ischemic preconditioning [1, 2]. Activation of mitochondrial permeability transition (mPT) has been demonstrated to play a critical role in several animal models of disease including ischemia-reperfusion injury, and inhibition of the mPT regulator cyclophilin D is cardioprotective in animal models and in man [3–6]. The mechanisms involved in preconditioning are not fully established but inhibition of mPT is generally regarded as an important end effect [7, 8].

Changes in mitochondrial H⁺ or K⁺ conductance may critically influence mitochondrial function. Several different pathways involving altered H⁺ or K⁺ entry into the mitochondrial matrix have been proposed to modify resistance to ischemia-reperfusion injury. Mitochondrial K⁺ channel activation, in particular the putative ATP-sensitive K⁺ channel (MitoK_ATP), has been extensively investigated as a mediator of ischemic preconditioning [9]. Uncoupling proteins - i.e. mitochondrial transporter proteins structurally related to UCP1, which uncouples mitochondrial respiration in brown adipose tissue - have also been explored as mediators of ischemic preconditioning [10]. Activation of K⁺ channels and H⁺ translocators will both uncouple mitochondrial oxidative phosphorylation to some extent. So-called mild uncoupling [11] has been suggested to reduce the sensitivity of mitochondria to undergo calcium-induced mPT both by reducing the ∆Ψ_m-dependent calcium uptake and by attenuating reactive oxygen species (ROS) production of mitochondria [12, 13]. Low concentrations of protonophores can induce a ROS-dependent cardioprotection in ischemia-reperfusion injury [14]. Similarly, activation of mitochondrial K⁺ channels has been proposed to reduce the generation of ROS [15, 16]. However, results are not consistent as the opposite relation also has been found with an increase in mitochondrial ROS production following K⁺ channel activation [17]. A signaling pathway linking mitoK_ATP and mPT has also been proposed where an increased generation of H₂O₂ activates an mPT-associated PKCϵ [18, 19]. The increased K⁺ conductance following mitochondrial K⁺ channel activation may also increase matrix volume [20]. We have previously demonstrated that enhanced matrix volume in brain mitochondria increased their resistance to calcium-induced mPT [21].

The objective of the present study was to explore how increased H⁺ and K⁺ conductance influence a number of heart mitochondrial physiological parameters relevant for ischemia-reperfusion injury, including respiration, calcium retention capacity (CRC), matrix volume, ROS production and ROS scavenging. Further, we wanted to explore whether increased matrix volume in heart mitochondria exert a beneficial effect on CRC.

In summary, we demonstrate that decreased matrix volume reduces mitochondrial resistance to calcium-induced mPT whereas increased volume did not alter it. H₂O₂ generation of mitochondria was reduced by both K⁺- and H⁺-mediated respiratory uncoupling whereas scavenging of H₂O₂, which was substantially greater than the production rate, was unaffected by uncoupling unless the inner mitochondrial membrane integrity was disrupted. Increased H⁺ or K⁺ flux dose-dependently decreased mitochondrial calcium retention capacity and no level of uncoupling demonstrated any positive effect on heart mitochondrial resistance to calcium-induced mPT.

Several hypotheses on the mechanisms of ischemic preconditioning involve activation of channels or transporter proteins affecting the ion flux, most notably for K⁺ and H⁺, across the inner mitochondrial membrane. Increasing mitochondrial K⁺ or H⁺ conductance will have several physiological consequences. The aim of the present work was to explore how some of these alterations relate to commonly proposed mechanisms for ischemia-reperfusion injury in heart mitochondria.

An extensive uncoupling will cause a dramatic effect on normal mitochondrial function, but so-called mild uncoupling has been suggested to reduce the sensitivity of mitochondria to undergo calcium-induced mPT by limiting the calcium uptake and ROS production of mitochondria [12, 25]. The CRC assay is a sensitive and quantitative method for evaluating mitochondrial susceptibility to mPT induction [26], and by simultaneous monitoring of respiration the specificity is increased [21]. Here, we demonstrate that heart mitochondrial CRC is dose-dependently reduced by uncoupling irrespective of whether H⁺ or K⁺ conductance is increased, and that the reduced CRC is caused by mPT (as demonstrated by the respiratory decrease). There was no tendency of any beneficial effect at lower concentrations of the K⁺ carrier valinomycin or the protonophore CCCP. At the highest tested concentration of CCCP, 200 nM, the respiratory increase was less than a state 2 to state 3 transition but the ability to retain Ca²⁺ was virtually abolished and mitochondria underwent permeability transition at a very low calcium load. Ca²⁺ is the principal trigger of mPT and a reduction in matrix Ca²⁺ would arguably reduce mPT activation. However, it is the free Ca²⁺ ions that trigger mPT and most calcium in mitochondria is retained as inactive calcium-phosphate complexes [26]. The formation of calcium-phosphates is highly dependent on an alkaline matrix environment. A small decrease in ∆pH, as occurs with uncoupling following H⁺ entry, will increase the concentration of free Ca²⁺ even if the electrophoretic driving force for Ca²⁺ uptake is reduced. Uncoupling may also directly increase the propensity of the mPT pore to open due to a more oxidized redox state and a reduced mitochondrial membrane potential (∆Ψ_m) [27, 28].

Increasing K⁺ flux into the mitochondria may affect several factors important for calcium handling and mPT regulation, including ∆Ψ_m, ∆pH, ROS generation and matrix volume, which are somewhat different to increasing H⁺ flux. For example, increasing K⁺ or H⁺ conductance may be expected to have opposite effects on ∆pH due to activation of the K⁺/H⁺ exchanger by the former. Whether an alteration in any of these parameters is beneficial or not to the resistance of mitochondria to the dramatic consequences of calcium overload depend both on their direct and potential indirect effects on the mPT pore components, as pointed out above for the effect of ∆pH on free Ca²⁺. In a previous study, we found that low increases in K⁺ but not H⁺ conductance was beneficial to brain mitochondrial resistance to calcium-induced permeability transition, and that the beneficial effect probably was mediated by an increase in matrix volume [21]. In the present study, there was no qualitative difference between increasing H⁺ or K⁺ conductance on CRC. The negative effects of uncoupling thus predominate over other potential direct beneficial effects of increasing K⁺ conductance in rat heart mitochondria.

Changes in mitochondrial matrix volume influence respiratory function [29–31] and we have previously demonstrated that increased matrix volume in brain mitochondria either caused by an elevated influx of K⁺ or by suspending mitochondria in hypoosmolar KCl medium increases CRC [21]. In the present study we explored respiratory changes and CRC in heart mitochondria suspended in hypo- or hyperosmolar KCl media. Increasing medium osmolarity by KCl (or LiCl) dose-dependently decreased CRC indicating a positive correlation between matrix volume and CRC. A possible confounder to increasing osmolarity with KCl or LiCl is the simultaneous change in ionic strength which may affect e.g. cytochrome c charge interactions [32]. However, increasing medium osmolarity with sucrose likewise decreased CRC. In contrast to the effects of increasing medium osmolarity, decreasing the osmolarity further than the standard conditions did not cause the expected increase in CRC. A plausible explanation may be the reduced respiratory capacity of heart mitochondria in the hypoosmolar medium. An increase in matrix volume has been shown to cause release of cytochrome c thus reducing respiratory capacity in isolated liver mitochondria [33]. The same authors also found a reduced calcium uptake in mitochondria following exposure to a hypoosmotic shock. In a limited set of experiments in heart mitochondria (data not shown) we found that there was a tendency for exogenously added cytochrome c to increase respiration more in hypoosmolar media compared to hyperosmolar conditions, but these findings were not explored further. It is thus plausible that any beneficial effect on heart mitochondrial CRC by an enlarged matrix volume following increased K⁺ flux is counteracted by a respiratory inhibition and a direct negative effect of uncoupling. Even though the majority of studied characteristics described herein were similar to previous findings in brain mitochondria, there seems to be a distinct organ specific difference in relation to the response to increased K⁺ conductance and matrix volume.

ROS production of mitochondria has since long been recognized to increase with elevated proton-motive force [34, 35]. This is particularly evident when isolated mitochondria are supplemented with solely succinate as respiratory substrate to support so-called reverse electron transport through respiratory complex I [36]. Ischemic preconditioning reduces oxidative damage in mitochondria, and a reduction of oxidative stress may mediate a reduced propensity for mPT activation [37]. Previous studies have demonstrated both increased and decreased ROS generation following augmented K⁺ flux in mitochondria [15–19]. In the present study, both increased K⁺ and H⁺ flux dose-dependently reduced the production of H₂O₂ in heart mitochondria using complex I substrates. There was no apparent qualitative or quantitative difference between increased K⁺ and H⁺ flux; rather, the level of ROS reduction was in proportion to the extent of uncoupling. However, the reduced ROS generation did not translate into an increased resistance to mPT induction.

The capacity of mitochondria to detoxify ROS is less explored but likely equally or more important than mitochondrial ROS production in oxidative stress [23] and there is no clear evidence whether mitochondria function as net source or sink of ROS [38]. Heart mitochondria are equipped with effective antioxidant systems for scavenging of ROS and dysfunction of these are involved in cardiovascular disease [39]. Malfunction of mitochondrial thioredoxin reductase causes dilated cardiomyopathy in man [40] and aggravates ischemia-reperfusion injury in mice [41], while overexpression of glutaredoxin-2 reduces ischemia-reperfusion injury in mice [42]. The mitochondrial antioxidative systems are dependent on intact respiratory function. Reduced glutathione and NADPH provide antioxidant defences against H₂O₂ mediated damage through the action of e.g. the glutathione peroxidase family, peroxiredoxins and glutaredoxins. The reduction of GSH and thioredoxin reductase is dependent on NADPH generation from e.g. the proton motive force-driven transhydrogenase [43]. Although uncoupling of mitochondrial respiration reduces ROS production, any interference with ROS scavenging may result in an overall negative effect on mitochondrial ROS handling. Therefore, we investigated the heart mitochondrial capacity of H₂O₂ removal under conditions of uncoupling and calcium overload. The mitochondrial H₂O₂ scavenging rate was more than 10-fold greater than the detected production rate of H₂O₂ under standard conditions. Whereas uncoupling decreased H₂O₂ production it did not affect H₂O₂ scavenging. In contrast, calcium overload leading to mPT dramatically reduced H₂O₂ scavenging in line with previous reports [23, 24]. Any effects of calcium unrelated to mPT activation were not explored in the present study.

We did not detect any specific effect of the MitoK_ATP opener diazoxide. The compound uncoupled respiration at high concentrations, which also has been demonstrated previously [20]. The present experiments were not primarily designed to explore the putative MitoK_ATP. Such experiments are usually performed within the first few minutes upon incubation in experimental medium and the present experiments examining general effects of increased H⁺ and K⁺ conductance required a longer time frame. It has been noted that the MitoK_ATP is sensitive to rapid inactivation [44] and there is also a current debate over its existence [9].

The present findings suggest that there are no direct beneficial effects of increased K⁺ or H⁺ flux on heart mitochondrial resistance to mPT. Even though there was a decrease in the generation of ROS and a positive correlation between matrix volume and CRC in hyperosmotic media, the negative effects related to respiratory uncoupling predominated with a net sensitization of heart mitochondria to mPT activation. Importantly though, uncoupling did not hamper mitochondrial ROS scavenging capacity. It is concluded that respiratory uncoupling is not beneficial to rat heart mitochondrial resistance to calcium overload regardless of whether H⁺ or K⁺ conductance is increased. The present study is limited by its specific examination of mitochondria in their isolated state in contrast to their integrated function in vivo. Nevertheless, the results imply that the direct physiological effects of increased H⁺ or K⁺ flux on ROS generation, matrix volume and calcium handling in heart mitochondria are not in themselves sufficient to mediate ischemic preconditioning in the heart.